Statistiek 1-5

Question 1

5 clean data checks

Answer 1

• Missing
• Range checks
• Outlier checks
• Distribution checks
• Logical checks

Question 2

verschil disreet en contious

Answer 2

discreet alleen hele getallen. continous kan alle getallen aannemen.

Question 3

Nominaal vs ordinaal?

Answer 3

nominal: > 2 categorieen niet geordenend: bloedgroep
ordinal: > 2 categorieen wel geordend: pijnscore 1-5

Question 4

mode, mean, median

Answer 4

Modus: getal wat meest voor komt. (meerdere of geen modus mogelijk

(arithmetic) mean: disproportionately affected by outliers. distributie checken van tevoren
median: center of the data set (ranked highest to lowest. less effected by outliers

Question 5

def: variance, def SD

Answer 5

Variance: mean of squared deviatons
SD: wortel(variance)

Question 6

Voordeel: mean + sd?

Answer 6

gebruikt alle informatie, makkelijk te inetrpreteren. cave: gevoelig voor outliers

Question 7

Voor/nadeel van assymetrische verdelingen standaaarden

Answer 7

Range: makkelijk e verrihten, easily distorted
Percentiles: niet zo snel distorted by outliers

Question 8

5 eigenscappen normale verdeling

Answer 8

• bell shaped
• symmetrical around its mean
• mean and median are equal
• one top
• fully described by mean and SD

Question 9

3 manieren van normaliteits bepaling

Answer 9

summary statisstics

visual inspection

formal tests

Question 10

Q-Q plot?

Answer 10

shows quantiles. X as = quantilen eigen data, Y as = verwacht als normale distributie gevolgd zou worden

Question 11

Downsides Kolmogorov Smrinov test and shapiro wilk test

Answer 11

• in small samples insufficiient power

- large samples, smal deviates are flagged as significantly deviant